[dhcwg] Re: udhcpd Win98 interoperability

> > define 'ours' a little better, do you mean if there is a static mapping 
> > for the address that we take care of? 
> >  
>  
> OK, I would say the requested address should be NAKed if 
> - we have static mapping for it on another client 
> - it is already assigned to one of our other clients 
> - it was assigned to the requesting client, but the lease has already timed out 
>  
> Do you think this is reasonable? 
>   

I'm not sure if nak'ing a lease that timed out is the correct behavior
if the address is still available, the lease can just be extended as if
it never ran out (apparent disagreement on system clocks). Here is the
new behavior my server has if it has no record of the client:

lease = find_lease_by_chaddr(packet.chaddr);

...

if (lease) {

...

/* what to do if we have no record of the client */
} else if (server_id) {
	/* SELECTING State */

} else if (requested) {
	/* INIT-REBOOT State */
	if ((lease = find_lease_by_yiaddr(requested_align))) {
		if (lease_expired(lease)) {
			/* probably best if we drop this lease */
                	memset(lease->chaddr, 0, 16);
		/* make some contention for this address */
		} else sendNAK(&packet);
	} else if (requested_align < server_config.start ||
		   requested_align > server_config.end) {
		sendNAK(&packet);
	} /* else remain silent */

} else {
	/* RENEWING or REBINDING State */
}

note: code is under the terms of the GPL

_______________________________________________
dhcwg mailing list
dhcwg@ietf.org
https://www1.ietf.org/mailman/listinfo/dhcwg